Weavers / MCP documentation

Preview a dataset

Inspect the schema identifier and a small row sample from an owned dataset version.

Read onlyget_python_dataset_preview

Required permission: weavers.datasets.preview

Try asking

Show five sample rows and the schema identifier for this dataset version.

How it works

The default is five rows and the maximum is ten. There is no pagination to download the dataset.
The assistant receives private sample rows. Grant the dataset-preview permission only when needed.
Use datasetVersionId from get_strategy_project. The sample is not a complete experiment input.

Inputs

Field	Required	Type & constraints
strategyId	Required	UUID string
datasetVersionId	Required	UUID string
limit	Optional	integer; minimum 1; maximum 10; default 5

View complete input JSON schema

Input schema

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "strategyId": {
      "type": "string",
      "format": "uuid",
      "pattern": "^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$"
    },
    "datasetVersionId": {
      "type": "string",
      "format": "uuid",
      "pattern": "^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$"
    },
    "limit": {
      "default": 5,
      "type": "integer",
      "minimum": 1,
      "maximum": 10
    }
  },
  "required": [
    "strategyId",
    "datasetVersionId"
  ],
  "additionalProperties": false
}

What comes back

Field	Description
datasetVersionId, rawSha256	Strings. Immutable dataset identity and hash.
schema	Schema identifier string: "weavers.python.daily_ohlcv.v1".
rowCount	Integer. Total dataset row count, not the sample size.
sampleOnly	Always true.
rows[]	Sample row objects using the dataset's normalized columns.

This table describes the decoded tool result. MCP wraps it in text content and, for object results, structuredContent.

Worked example

Fictional data. Replace identifiers, revisions and hashes with values from your own Weavers workflow.

Request

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_python_dataset_preview",
    "arguments": {
      "strategyId": "11111111-1111-4111-8111-111111111111",
      "datasetVersionId": "22222222-2222-4222-8222-222222222222",
      "limit": 5
    }
  }
}

Decoded result

{
  "datasetVersionId": "22222222-2222-4222-8222-222222222222",
  "rawSha256": "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
  "schema": "weavers.python.daily_ohlcv.v1",
  "rowCount": 120,
  "sampleOnly": true,
  "rows": [
    {
      "symbol": "EXAMPLE",
      "timestamp": "2026-01-15",
      "open": 99,
      "high": 102,
      "low": 98,
      "close": 100,
      "volume": 1000
    }
  ]
}

If the request fails

Sign in and grant weavers.datasets.preview before calling this tool.
Use resources you can access and refresh revision information when it changes.
Check result.isError even when HTTP returns 200.

Find the error and next step →